PREDIKSI TINGKAT DEPRESI MAHASISWA MENGGUNAKAN MODEL HYBRID XGBOOST DAN SVM BERDASARKAN PHQ-9, IPK, TAHUN AKADEMIK, DAN USIA

Ananda Kallila; Muhamad Bahrul Ulum

doi:10.23960/jitet.v14i2.9307

Ananda Kallila
Universitas Esa Unggul
Muhamad Bahrul Ulum

DOI: https://doi.org/10.23960/jitet.v14i2.9307

Keywords Depresi, XGBoost, Support Vector Machine, PHQ-9, Hybrid

Abstract Views (Last 12 Months)

135 Abstract Views

92 Downloads

Abstract

Kesehatan mental mahasiswa menjadi perhatian penting karena tekanan akademik dan sosial yang tinggi dapat memicu gejala depresi. Penelitian ini bertujuan membangun model klasifikasi tingkat depresi berdasarkan kuesioner PHQ-9 dan data demografis seperti usia, IPK, dan tahun akademik. Tiga pendekatan digunakan: Support Vector Machine (SVM), Extreme Gradient Boosting (XGBoost), dan model hybrid. Tahapan penelitian mencakup preprocessing data (imputasi, encoding, normalisasi), pelatihan model, dan evaluasi dengan akurasi, precision, recall, F1-score, dan AUC. Model hybrid dibentuk dengan menggunakan output probabilitas XGBoost sebagai input pelatihan model SVM. Hasil evaluasi menunjukkan SVM memiliki akurasi tertinggi sebesar 96%, diikuti hybrid 93,5% dan XGBoost 93%. Namun, pada pengujian data uji individual, model hybrid menunjukkan prediksi yang lebih tepat dibanding SVM. Seluruh model diimplementasikan dalam antarmuka berbasis web agar hasil klasifikasi mudah diakses oleh pengguna. Penelitian ini menunjukkan bahwa baik model tunggal maupun hybrid memiliki potensi besar dalam mengklasifikasikan tingkat depresi secara non-diagnostik. SVM unggul dalam klasifikasi makro, XGBoost efektif untuk pola fitur kompleks, dan pendekatan hybrid menawarkan solusi prediksi yang lebih stabil dan adaptif terhadap variasi data mahasiswa.

Downloads

Download data is not yet available.

References

C. F. P. S. A. Daza Vergaray, J. C. H. Miranda, J. B. Cornelio, A. R. López Carranza, “Predicting depression in university students using stacking ensemble techniques over oversampling method,” Informatics Med. Unlocked, vol. 41, 2023, doi: doi: 10.1016/j.imu.2023.101295.

N. I. F. Kaligis, R. Indrayanti, “Mental health problems and needs among transitional age youth in Indonesia,” Int. J. Environ. Res. Public Health, vol. 18, no. 6, 2021, doi: doi: 10.3390/ijerph18062912.

R. C. A. A. Khan, O. Chaudhari, “A review of ensemble learning and data augmentation models for class imbalanced problems: Combination, implementation and evaluation,” Expert Syst. Appl., vol. 244, 2024, doi: doi: 10.1016/j.eswa.2023.122778.

S. C. B. R. Krebs, S. S. Bagui, D. Mink, “Applying multi-class support vector machines: One-vs.-One vs. One-vs.-All on the UWF-ZeekDataFall22 dataset,” Electronics, vol. 13, no. 19, 2024, doi: doi: 10.3390/electronics13193916.

A. M. S. Dalal, E. M. Onyema, “Hybrid XGBoost model with hyperparameter tuning for prediction of liver disease with better accuracy,” World J. Gastroenterol., vol. 28, no. 46, pp. 6551–6563, 2022, doi: doi: 10.3748/wjg.v28.i46.6551.

Y. S. P. Zhang, Y. Jia, “Research and application of XGBoost in imbalanced data,” Int. J. Distrib. Sens. Networks, vol. 18, no. 6, 2022, doi: doi: 10.1177/15501329221106935.

H. H. and A. Idri, “A comparative study on the impact of categorical encoding on black box model interpretability,” Proc. 13th Int. Conf. Data Sci. Technol. Appl. (DATA 2024), pp. 384–391, 2024, doi: 384–391, doi: 10.5220/0012766300003756.

H. R. A. M. Razavi, S. T. Yazdani, H. R. Rabiee, “Machine learning, deep learning and data preprocessing techniques for detection, prediction, and monitoring of stress and stress-related mental disorders: A scoping review,” JMIR Ment. Heal., vol. 11, pp. 1–28, 2023, doi: doi: 10.2196/53714.

O. T. T. Emmanuel, T. Maupong, D. Mpoeleng, T. Semong, B. Mphago, “A survey on missing data in machine learning,” J. Big Data, vol. 8, no. 1, 2021, doi: doi: 10.1186/s40537-021-00516-9.

S. Schwartz, S. Ross, J. Bryant, and J. Duncan, “Mental and physical health among students at a private university that held in-person classes during the COVID-19 pandemic,” J. Am. Coll. Heal., vol. 72, pp. 1–11, Mar. 2022, doi: 10.1080/07448481.2022.2052074.

R. A. Saputra, “ANALISIS SENTIMEN REVIEW SKINCARE SKINTIFIC DENGAN ALGORITMA SUPPORT VECTOR MACHINE,” vol. 12, no. 2, 2024.

J. B. P. A. Perez-Lebel, G. Varoquaux, M. Le Morvan, J. Josse, “Benchmarking missing-values approaches for predictive models on health databases,” Gigascience, vol. 11, pp. 1–22, 2022, doi: doi: 10.1093/gigascience/giac013.

J. M. M. A. Rahman, T. A. Dhira, A. R. Sarker, “Validity and reliability of the Patient Health Questionnaire scale (PHQ-9) among university students of Bangladesh,” PLoS One, vol. 17, no. 6, pp. 1–18, 2022, doi: doi: 10.1371/journal.pone.0269634.

A. C. D.-B. C. A. Cassiani-Miranda, O. Scoppetta, M. A. Barrios-Villadiego, A. F. Tirado-Otálvaro, “Construct validity of the PHQ-9 in university students in Colombia: A Rasch analysis approach,” Med. Clínica y Soc., vol. 9, no. 1, 2025, doi: doi: 10.52379/mcs.v9.521.

X. Q. and X. Huang, “Machine learning-driven identification of key risk factors for predicting depression among nurses,” BMC Nurs., vol. 24, no. 1, 2025, doi: doi: 10.1186/s12912-025-02957-6.

U. A. and P. Agarwal, “DepXGBoot: Depression detection using a robust tuned extreme gradient boosting model generator,” IAES Int. J. Artif. Intell., vol. 13, no. 4, pp. 4352–4363, 2024, doi: doi: 10.11591/ijai.v13.i4.pp4352-4363.

N. D. R. Pranckeviciene, E. T. Leskauskas, M. Adomaitiene, “Validation of the PHQ-9 and GAD-7 in primary care and psychiatry settings,” BMC Psychiatry, vol. 22, no. 1, pp. 1–10, 2022, doi: doi: 10.1186/s12888-022-04452-5.

H. A. R. Ibrahum, A. Rahman, M. H. Ali, “Extreme Gradient Boosting-based prediction for depression among university students,” Int. J. Adv. Comput. Sci. Appl., vol. 14, no. 5, pp. 725–732, 2023, doi: doi: 10.14569/IJACSA.2023.0140588.

T. D. N. N. Vu, A. Q. Nguyen, “Prediction of depressive disorder using machine learning approaches,” IEEE Access, vol. 13, pp. 11987–11999, 2025, doi: doi: 10.1109/ACCESS.2025.3500142.

R. K. T. S. Singh, R. K. Saini, A. Sharma, “Machine learning algorithms for detecting mental stress: A comprehensive review,” Artif. Intell. Med., vol. 144, p. 102658, 2024, doi: doi: 10.1016/j.artmed.2024.102658.

K. K. R. Indonesia, “Survei Nasional Kesehatan Mental Remaja Indonesia (I-NAMHS): Laporan Nasional,” Kementerian Kesehatan RI.

W. H. Organization, “World Health Organization, ‘Depressive disorder (depression),.’” [Online]. Available: https://www.who.int/news-room/fact-sheets/detail/depression

O. V Chukwuma et al., “A Systematic Review of the Effect of Stigmatization on Psychiatric Illness Outcomes.,” Cureus, vol. 16, no. 6, p. e62642, Jun. 2024, doi: 10.7759/cureus.62642.

L. Guo et al., “Prevalence and changes in depressive symptoms among postgraduate students: A systematic review and meta-analysis from 1980 to 2020.,” Stress Heal. J. Int. Soc. Investig. Stress, vol. 37, no. 5, pp. 835–847, Dec. 2021, doi: 10.1002/smi.3045.